Ranjit Mathew wrote:
Hi,
Between Tuesday and Wednesday (Indian time), something(s)
went into mainline that is showing me a dramatic decrease
in bootstrap times - a c,c++,java bootstrap on i686-pc-linux-gnu
now takes 51m for me compared to 65-66m earlier, which is
around 20% of savings over the co
On 5/19/05, Paolo Bonzini <[EMAIL PROTECTED]> wrote:
> Ranjit Mathew wrote:
> > Hi,
> >
> > Between Tuesday and Wednesday (Indian time), something(s)
> > went into mainline that is showing me a dramatic decrease
> > in bootstrap times - a c,c++,java bootstrap on i686-pc-linux-gnu
> > now takes 51
Mike Stump wrote:
> mrs bash[73] nm i586-pc-linux-gnu/libobjc/.libs/libobjc.so.1 | grep
> gcc_unre
> U gcc_unreachable
>
> :-(
>
> This is killing the Objective-C testsuite for me...
nm -o *.o |grep gcc_unr
exception.o:U gcc_unreachable
It seems this comes from unwind.h
static
Hello!
A few LINK_SPEC definitions contain a "%{Wl,*:%*}" sequence. AFAICT this
sequence doesn't have any effect since at least GCC 2.95, because gcc.c
doesn't recored "-Wl" options in array "switch" (used in process_brace_body)
but stores them in "infiles". Should it be possible to match against
Recently I've been reading about General-Purpose Computation Using
Graphics Hardware - http://www.gpgpu.org - and it seems that GPUs can
bring quite a good performance when compared to the CPUs.
In other words, a graphics chip on the graphics card can make really
heavy computations, and it's ea
I have been building gcc-4.0.0 from Apple sources with tags in the
apple-ppc-5000 series.
I was getting lots of messages like this "spec failure: unrecognized
spec option 'Q'"
I notice that APPLE LOCAL tagged changes seem to turn spec warnings
on. Presumably they are off by default.
I ame
On 5/19/05, Paolo Bonzini <[EMAIL PROTECTED]> wrote:
>
> If you want, you can help by testing the patch for PR17485 against
> jacks; since a few days, I cannot get jacks to run (but I think it's my
> fault, not the tree's).
FWIW, I tested your patch for PR17845 by doing a clean
bootstrap and run
On 5/19/05, Ranjit Mathew <[EMAIL PROTECTED]> wrote:
[...]
> FWIW, I tested your patch for PR17845 by doing a clean
> bootstrap and running the entire libjava testsuite (including
> Jacks) and found no regressions on i686-pc-linux-gnu.
By "clean", I meant a CVS snapshot with your patch
applied and
I met a strange problem when using a .a file built by arm-linux-gcc
2.95.3 with arm-linux-gcc-3.4.3:
[EMAIL PROTECTED] aac]# arm-linux-ld -r -o ../lib/audAACdec.a
_lib/mpglib.a build/aactbl.o build/aacdec.o build/aacinit.o
build/aacutil.o build/mp4aacHeader.o
This works well for me but move _li
> > Jan Hubicka writes:
>
> >> That might be related to the bootstrap failure on AIX as well.
>
> Jan> Hopefully this is fixed now by Jeff's patch.
>
> The libjava failure is fixed, but the patch will not affect the
> AIX libgfortran failure.
>
> I have verified that either the
On Wed, 18 May 2005 17:26:30 +0200, Marcin Dalecki wrote:
> Like building on the system you are targeting?
> Like cross building for the target system?
No, like messing around with headers and linkers and compilers, so if you
are targetting Linux/x86 your binaries can in fact run on Linux/x86.
I
On Wed, 18 May 2005 14:29:12 -0700, Joe Buck wrote:
> These kinds of problems can be solved, but they are beyond the scope of
> this list.
Well, alright - but which list is it in scope for? GCC is generally pretty
good about this but in a few parts where it interacts with glibc like the
EH code, v
On Wed, 18 May 2005 22:02:37 +0100, Paul Brook wrote:
> Rubbish. You've obviously never tried to install two third party windows
> applications that require two different revisions of msvcrt.dll, or even
> worse two random versions of an OCX control.
And Linux has dependency hell.
As a long ti
Hi,
There is a typo in french translation of error messages (at least in 4.0
release and in 4.1 snapshot of 05/05/15).
It affects gcc/po/fr.po and libcpp/po/fr.po :
"sasn effet" should be "sans effet"
"sasn lien" should be "sans lien"
(The word "sasn" does not exist in french language).
Regards,
> "DLL hell" is a Windows term, not a Unix/BSD/GNU/Linux term, for a
reason.
Unix would have "SO woes".
Paolo
On 2005-05-19, at 15:18, Mike Hearn wrote:
On Wed, 18 May 2005 17:26:30 +0200, Marcin Dalecki wrote:
Like building on the system you are targeting?
Like cross building for the target system?
No, like messing around with headers and linkers and compilers, so
if you
are targetting Linux/x86 your bi
I'd need some help to develop a function, for a new gcc pass,
located after the value range propagation pass.
I'm working on the last release of gcc 4.1.
I'd like to have a function that receiving a stmt (that is
a CALL_EXPR) and an index (to specify the position of the
argument), has to return,
On Thu, May 19, 2005 at 02:22:42PM +0100, Mike Hearn wrote:
> On Wed, 18 May 2005 22:02:37 +0100, Paul Brook wrote:
> > Rubbish. You've obviously never tried to install two third party windows
> > applications that require two different revisions of msvcrt.dll, or even
> > worse two random versi
On Thu, 2005-05-19 at 16:42 +0200, Davide Pozza wrote:
> I'd need some help to develop a function, for a new gcc pass,
> located after the value range propagation pass.
>
> I'm working on the last release of gcc 4.1.
>
> I'd like to have a function that receiving a stmt (that is
> a CALL_EXPR) an
On Thu, May 19, 2005 at 04:42:29PM +0200, Davide Pozza wrote:
> Calling the get_def_operand for this statement with index 0, the
> function has to return the tree corresponding to buf_13.
>
If the argument you are looking for is not call-clobbered (not in
the call_clobbered_vars set) you can retu
On Thu, 19 May 2005 16:38:12 +0200, Marcin Dalecki wrote:
> Ah I see. You live under the perhaps marketing inflicted self delusion
> that Linux/x86 denominates a single well defined system.
> Welcome to the reality then.
99% of the time, it does. The fact that in a few areas it doesn't is
somethin
I'm trying to match up call arguments with *_DEF's.
My aim is to build a taint analysis engine to track (along the code)
what variables are under user control (i.e. to track user input
variables obtained by means of functions along the code).
For example, in the case of an fgets, the first parame
Richard Henderson wrote:
After three days of sequential bootstrap breakage, I'd like to propose
that mainline go into slush, wherein all these bootstrap problems, and
all the new testsuites failures get fixed. No other patches would be
allowed at all.
We'd unslush when the primary platforms have c
> If you're running tests on a primary platform, and think things are
> OK, please send me an email pointing at gcc-testresults mail showing
> allegedly clean results for that platform *and* update:
>
> http://gcc.gnu.org/wiki/GCC%204.1%20Slush
I ran mainline tests checked out last night on
On Thu, May 19, 2005 at 10:11:54AM -0700, Mark Mitchell wrote:
> If you're running tests on a primary platform, and think things are OK,
> please send me an email pointing at gcc-testresults mail showing
> allegedly clean results for that platform *and* update:
>
> http://gcc.gnu.org/wiki/GCC%2
On May 19, 2005, at 2:44 AM, Bill Northcott wrote:
I have been building gcc-4.0.0 from Apple sources with tags in the
apple-ppc-5000 series.
I was getting lots of messages like this "spec failure:
unrecognized spec option 'Q'"
I amended /usr/lib/gcc/powerpc-apple-darwin8/4.0.0/specs
rm -rf /u
On May 19, 2005, at 1:31 PM, Kaveh R. Ghazi wrote:
If you're running tests on a primary platform, and think things are
OK, please send me an email pointing at gcc-testresults mail showing
allegedly clean results for that platform *and* update:
http://gcc.gnu.org/wiki/GCC%204.1%20Slush
I ran mainli
On May 19, 2005, at 10:11 AM, Mark Mitchell wrote:
Nobody's objected, and it's fine by me. So, let's do it.
Ping.
I kinda wish someone would review the libjava breakage patch for
darwin...
http://gcc.gnu.org/ml/gcc-patches/2005-05/msg01821.html
otherwise, I don't see the point in slushing to fi
On Thu, May 19, 2005 at 02:28:46PM +0100, Mike Hearn wrote:
> On Wed, 18 May 2005 14:29:12 -0700, Joe Buck wrote:
> > These kinds of problems can be solved, but they are beyond the scope of
> > this list.
>
> Well, alright - but which list is it in scope for? GCC is generally pretty
> good about t
On Thu, May 19, 2005 at 04:42:42PM +0100, Mike Hearn wrote:
> On Thu, 19 May 2005 16:38:12 +0200, Marcin Dalecki wrote:
> > Ah I see. You live under the perhaps marketing inflicted self delusion
> > that Linux/x86 denominates a single well defined system.
> > Welcome to the reality then.
>
> 99% o
On May 19, 2005, at 2:13 PM, Mike Stump wrote:
On May 19, 2005, at 10:11 AM, Mark Mitchell wrote:
Nobody's objected, and it's fine by me. So, let's do it.
Ping.
I kinda wish someone would review the libjava breakage patch for
darwin...
This was already applied though not with your name on the cha
On Thu, 19 May 2005, Paolo Bonzini wrote:
> I do not know if it is entirely because of the patch to build libjava
> one-directory-at-a-time. If it is, you should thank me :-) for the
> patch, and rth for reviewing it very quickly.
In any case, I'll note that you fixed the libjava build on FreeB
Perhaps sending this to java-patches will help...
Mike Stump wrote:
On May 19, 2005, at 10:11 AM, Mark Mitchell wrote:
Nobody's objected, and it's fine by me. So, let's do it.
Ping.
I kinda wish someone would review the libjava breakage patch for darwin...
http://gcc.gnu.org/ml/gcc-patches/2005-
"Joseph S. Myers" <[EMAIL PROTECTED]> writes:
> On Wed, 18 May 2005, Richard Sandiford wrote:
>> FWIW, those mips messages _were_ sent with test_summary. I'm not
>> sure why they got mangled. Are there any known issues that would
>> cause that?
>
> What awk version are you using? (The script tes
I'm porting gcc to a uni-core architecture (i.e., only one core).
There are 10 function units:
(1) 2 RISCs: the 2 RISC have the same capability and they can do load/store, full-word arithmetic/logic
operations, register move, ...
(2) 4 DSPs ( 2 MAC, 1 BSU, and 1 VFU):
Mark Mitchell <[EMAIL PROTECTED]> writes:
> Richard Henderson wrote:
>> After three days of sequential bootstrap breakage, I'd like to propose
>> that mainline go into slush, wherein all these bootstrap problems, and
>> all the new testsuites failures get fixed. No other patches would be
>> allowe
Ling-hua Tseng wrote:
>
> For example:
> ===[top]
> mov .risc0 r1, #25 \\
> ldw .risc0 r2, [fp, #30] \\
> addub .mac0 d0, d4, d3 \\
> subub .mac1 d11, d7, d4
> add .risc0 r3, r1, r5
> ===[end]==
Was this not fixed by:
2005-05-18 Paolo Bonzini <[EMAIL PROTECTED]>
* Makefile.am (Makefile.deps): Do not use \0, it is unportable.
* Makefile.in: Regenerate.
?
Bryce
David Daney wrote:
Perhaps sending this to java-patches will help...
Mike Stump wrote:
On May 19, 2005, at 10:11 AM, Mark Mi
On May 19, 2005, at 2:53 PM, Bryce McKinlay wrote:
Was this not fixed by:
2005-05-18 Paolo Bonzini <[EMAIL PROTECTED]>
* Makefile.am (Makefile.deps): Do not use \0, it is unportable.
* Makefile.in: Regenerate.
?
Yes, he checked in my change, and didn't copy me on the email...
Also, somet
> Neither does Linux - by linking against a recent library you are
> *asking* for a binary that requires that library. If you understand
> that you might understand why everyone is saying you should build on the
> lowest common denominator of the systems you're targetting.
>
> If you insist on shi
I want to do a binding to ObjectiveC
Basically I need to implement a function like
retval_t objc_msg_send(id obj, SEL sel, ...);
which would forward its call (same arguments, including the unidentified ...
(va_list)) to an other function know by pointer which would be looked up
from the 1st 2 par
On May 19, 2005, at 4:08 PM, Lloyd Dupont wrote:
I want to do a binding to ObjectiveC
For how you described the question, libffi would be the natural
choice and obviates the need for asms or machine dependencies. Maybe
Andrew might have some insight into something libobjc specific that
might
On May 19, 2005, at 7:49 PM, Mike Stump wrote:
On May 19, 2005, at 4:08 PM, Lloyd Dupont wrote:
I want to do a binding to ObjectiveC
For how you described the question, libffi would be the natural choice
and obviates the need for asms or machine dependencies. Maybe Andrew
might have some insight
Simple, I'm building an ObjectiveC binding from .NET.
Basically my .NET code generator transform all objectiveC method in a call
to
objc_msg_send() (from the .NET code, which could call C directly)
the problem with _built_in_apply() is you need to know the size of the
stack.
and that's a real is
The problem with ffi and such (from the little I know of them) is that I
might have to forward function with an unknown number of parameter (such as
printf)
whereas a little longjmp would solve all my problem, wouldn't it?
- Original Message -
From: "Mike Stump" <[EMAIL PROTECTED]>
To:
On May 19, 2005, at 8:10 PM, Lloyd Dupont wrote:
Simple, I'm building an ObjectiveC binding from .NET.
Basically my .NET code generator transform all objectiveC method in a
call to
objc_msg_send() (from the .NET code, which could call C directly)
the problem with _built_in_apply() is you need to
Jonathan Wilson wrote:
On windows, it is possible to build a binary using a compiler on Windows
XP that can then run on older versions of windows simply by not using
any features specific to the newest versions of windows XP (or by using
LoadLibrary and GetProcAddress to see if those features ar
I think that current documentation of `-fstack-check' is unclear. The
documentation states that for single-treaded program `-fstack-check'
is not usefull. IMHO for main thread `-fstack-check' is harmfull
and may cause spurious segfault. Namely the following silly
program:
extern int printf(const c
On 20/05/2005, at 3:51 AM, Mike Stump wrote:
I have been building gcc-4.0.0 from Apple sources with tags in the
apple-ppc-5000 series.
I was getting lots of messages like this "spec failure:
unrecognized spec option 'Q'"
I amended /usr/lib/gcc/powerpc-apple-darwin8/4.0.0/specs
rm -rf /usr/lib
After lots of test, (and some experimental ASM which I quickly discard for
my too rusty ASM skills) I found this nice solution build from idea
from the sources:
(However an ASM version with a jump would have been much more reliable
(regarding argument forwarding, particularly with variable argum
Kriang Lerdsuwanakij wrote:
Mark Mitchell wrote:
OK. Do you happen to have access to any other testsuites, beyond the
GCC testsuite? If so, it would be great to validate the behavior of
the compiler on the 4.0 branch with and without your patch to make
sure that we're not doing any harm.
I a
On 2005-05-16, at 22:03, Mark Mitchell wrote:
Georg Bauhaus wrote:
On Mac OX X 10.2, the results are slightly discomforting,
even though I do get a compiler with
--enable-languages=c,ada,f77,c++,objc.
gcc summary has
# of unexpected failures1080
First, I would suggest disabling Ada, in ord
52 matches
Mail list logo
